This proposal is designed to prospectively examine the interaction between two types of endurance exercise and the course and outcome of pregnancy. Its goal is to test the broad hypothesis that, above a threshold level, endurance types of exercise have a negative, dose-response effect on multiple aspects of the reproductive process in female recreational endurance athletes. Data addressing potential mechanisms for this effect will be obtained by evaluating each subject's physiological response to exercise. Over 5 years the actual exercise performance and the cardiovascular, metabolic, respiratory, and thermal response to two types of endurance exercise will be evaluated in 150 well-conditioned women prior to and during pregnancy. 75 runners and 75 aerobic dancers form the study populace which will be prospectively matched with a control group. Subsets of these groups will be used to evaluate specific physiological responses in greater detail. Methodologies include: underwater weighing and skinfold measurements for assessing body composition; a daily exercise log and portable heart rate monitor for assessing actual exercise performance; treadmill testing with appropriate respiratory and thermal equipment for assessing VO2 max and the metabolic, respiratory and thermal response to exercise; menstrual record, Beta-subunit and ultrasound for accurate pregnancy dating, standard enzymati and RIA methodologies for measurement of blood levels of catecholamines, glucose, lactate, insulin and cortisol; and standard morphometric techniques for assessment of fetal, placental and neonatal growth. The physiological data and exercise log will be used to calculate both time-specific and cumulative exercise stress expressed as a duration-intensity index. This independent variable will be correlated with multiple reproductive outcome variables such as: fertility; early and late pregnancy complications; course of labor and delivery; neonatal and placental morphometrics; and, growth through the first year of life. Specific subsets of physiological data will be used to assess the potential mechanisms likely to be involved in the interaction. Currently, there is a paucity of information in this area and the information obtained will have direct application to 25% of women considering pregnancy.